[nikongod]

cobaltmute:
Thanks for telling me how to do that: Its been ages since I have read HTML code, but I guess its time to relearn that.
 
Quote:

Unfortunately, right now all I can do is calculator math gymnastics. I don't have the money yet for parts or testing equipment (I plan to get a cheap oscilloscope and signal generator even though I know that they TECHNICALLY aren't essential pieces of equipment) yet. It wont be until mid next month that I will have the money to start going in at this. So in the mean time it's calculate away while trying to grasp better the nuance of tube amp design. At this point, my weakest areas of knowledge are at the "ends", meaning input and output considerations and calculations, and tube selection itself.

This last bit is the problem.
 
Dont worry about the tubes. What you need to worry about to design a (at least half) decent amp of ANY kind is exactly what you dont know, thats OK, admitting you dont know it is how you go out and learn it.
 
Before you think about more schematics & topologies and whatnot you need to go out and figure out what it is you are trying to accomplish. That is the key. Are you trying to put 100v into a 300ohm load? Are you trying to put 2a into 32ohms? Do you want an output impedance less than 0.0001ohms? what *EXACTLY* are you trying to do? For sure you could design an amp that would accomplish these feats, but do you have any real need for that beyond mental masturbation and bragging rights?
 
The really good news is you probably have most of what you need to figure this out. the DV337 should be able to drive a sennheiser to dumb volumes. Measure the ac voltage that it takes (its much much better if your meter will accurately display less than 2vac). If your meter has AC current, do that to or just crunch the numbers. Test it out on a grado too. learn for yourself exactly what it takes to drive these headphones. people love to spout rules about how to drive a senn, How many people are just parrots repeating unconditional rules they dont understand?
 
Perhaps even more telling, and where many people stray (IMO) is what it takes to drive these headphones to your own listening levels (for this you will almost certainly need a meter that displays less than 2vac with good precision). Does it matter that someone else's amp can put 20v into 300ohms when you only use 0.25v? I would say hardly so, what matters is that YOUR amp sounds nice at realistic volumes. does this necessarily mean that an amp that only puts out 2.5v (0.25v+20db headroom) WILL sound better? not at all, but the odds are stacked in your favor.
 
Now would be as good of a time as any to figure out what output impedance you need. Guess what, its not going to be zero with tube output stages. If you need that (and maybe you do, although I have not found that to be the case) have fun with SS stuff & hybrids. Again figure out what you NEED. The DV amps have kind of high output impedances for testing, so build a cmoy with a 500ohm pot on the output to adjust output impedance. Try a bunch of things out with a bunch of headphones.
 
So far you have spent the cost of building a cmoy plus a cheap 500ohm pot (and a meter that can display less than 2vac, if you dont have that) and learned a bunch.
 
Quote:

I found a GREAT resource on //feed but how that interrelates with the design of the tube stage and accompanying passive elements I have seen little. When it comes to tube selection, I just don't know that many yet. There are soooo many so having an idea of what tubes offer the gain I am looking for along with the Vg I am looking for, along with the Iq I am looking for, and the grid/V swing I am looking for (if I go pentode top) is more just google searches for "low-mu medium grid" and/or luck of the draw. Not the most expedient of means to find such things but also pretty interesting in its own right.

Dont worry about the tubes. Worry about what you need out of the amp. 
 
After you have figured out what you need out of the amp you can work the rest backwards from the output. You will probably be surprised how well things fall into place.
 
Quote:

When it comes to using a pentode as the top tube, there are 2 things: finding a tube that has the same Iq as the lower triode (haven't looked too far into this so this may prove to be a non issue), and my 337 already has a pentode in it so I was hoping to go all triode and see how different it would sound =D

The amps are totally different, even with the same tubes the amps would sound and work differently. By not using a cathode follower (unless you really want to use a Mu follower, in which case you owe it to yourself to listen to a few mu follower amps to help you change your mind) you have already changed the sound of the amp a great deal. Pentodes make great CCS's, why not use one? There are a couple disadvantages in a parafeed amp: you need higher voltage rails than with a SS CCS and you need a second heater supply.

 

[burgunder]

Hey it's your project so you are the one who's in charge of how fast it's progessing, so just keep it up the work and I'm sure you will succeed

 

[regal]

NikonGod gave good advice.    My design that I am working on now was zero and I mean zero caps in the signal path all the way back to the powersupply,  so I worked from there.  It reminds me of my bosses always saying start with a vision statement and work from there.  Yes I think it is possible,  hope to know shortly.  I came to this after learning that capacitors highly influence the sound to my ears just by switching brands.
 
With a parafeed design you may also want to look at the newer gyrator circuits in place of a ccs,  reports are that they keep the tube more linear?  I haven't tried them but heard good reports and they are cheap to try.  Sorry if this has already been mentioned as the new site is very slow for me to navigate page to page.

 

[dBs]

Quote:

NikonGod gave good advice.    My design that I am working on now was zero and I mean zero caps in the signal path all the way back to the powersupply,  so I worked from there.  It reminds me of my bosses always saying start with a vision statement and work from there.  Yes I think it is possible,  hope to know shortly.  I came to this after learning that capacitors highly influence the sound to my ears just by switching brands.
 
With a parafeed design you may also want to look at the newer gyrator circuits in place of a ccs,  reports are that they keep the tube more linear?  I haven't tried them but heard good reports and they are cheap to try.  Sorry if this has already been mentioned as the new site is very slow for me to navigate page to page.

Tried a different browser to see if it's browser specific? I used to freelance web design in high school for spare cash (back when HTML was cutting edge) and back then I made use of iFrames and image maps for my personal sites and they weren't supported by all browsers so wonky things would happen in those instances.
 
(back to amp design) Thus far I have taken the tactic of designing around the amp itself with the goal of balancing the many various trade-offs, steering to the lowest noise and most linear possible, without resorting to really complex topologies. The idea is that, as far as the output was concerned, I would tailor the transformer to the output impedance, voltage, current that I wanted to fit the headphones I am looking to interface with. The method would be to take the output impedance, voltage, and current of my amp, find the best corresponding primary value, and decide on the best turns ratio to achieve the Z, V, and I that I wanted for my headphones. Thus, I haven't strongly concerned myself as much with those factors as I have seen them easily flexible given that relationship. I could be thinking about this in too much of an idealized "perfect" world though, and in real world amp design this tactic may not be affective. Is this a correct assessment?
 
Unfortunately, it was my lack of foresight that resulted in the input V/I/P problem. In all the things I've read, no one has mentioned anything about that consideration (probably assuming that it was implied, lol). The same has been the case for the output properties to aim for (in a generalized sense). Everyone is so focused on the sexy amp topologies and misc. tweaks that they overlook the input and output properties that matter in the design and end result. I'm not trying to make excuses, but I am honestly surprised that these problems were able to "sneak up" on me without my reading about the ranges (in the real world) that tend to be acceptable. Kind of thought these would be advertised specs on my amp or other market amps where I could find a generalized range to base from. Well, lesson learned. Hopefully this thread will help those who follow behind me. That is my biggest goal here and why I plan to have everything be known and open for all to see from the math to the design and the build, I want another beginner like me to be able to know every step in the process.
 
Yea, I honestly jumped the gun a bit. It has taken me a bit longer to pay off (long story XD) a rather large sum of debt I had accumulated due to my job. I will be at zero debt mid next month though (minus student loan and car), so I will be able to start buying equipment and parts. I am just determined to be paid off before I do anything else, I just didn't think it would take this bloody long XD

 

[dsavitsk]

It's a bit of  messy subject. Actual necessary gain and output power for headphone amps is really quite modest. IMO, gain for low Z phones probably should be unity or lower, and for high Z, more like 3 to 4. People will tell you higher, but then, those people rarely turn the volume knob past 10:00 -- it seems to give people a sense of "power" if the output ramps up quickly rather than over the whole range of the pot.  Moreover, output power likely doesn't need to be over 250mW with perhaps 400mW a reasonable headroom target.  And even that is high -- the 1st mW is what counts.  Indeed, amps using things like 300B's are necessarilly compromised -- one noted transformer manufacturer pointed out that a single feed 45 needs a core that that can normally handle 70W ungapped to handle the gap and DC necessary for that tube to produce under 2W.  All that does is throw away low level detail and render the tube very ordinary.  A 300B is worse.  Big power tubes sell amps, but they don't work better.
 
Not all phones are happy with that, but the bulk are, and trying to make an amp that pleases everyone will no doubt end up pleasing no one. For the esoteric phones, it is better to design something esoteric.
 

 

[FrankCooter]

As usual, I agree with about 99% of what Doug says. But in the case of his comment about big power tubes,  I'd like to point out that the Japanese DIY community, which historically has been the forefront of the DHT revival, has been running large transmitting tubes at low levels for years. However much this violates generally accepted engineering principals, the results can be surprizingly, even stunningly, good. Sorry to drift off point.
 
dBs, it's definitely time to end the "math gymnastics" and break out the soldering iron.Your making this a lot more complicated than it really is. Most of your problems will be in implementation rather than design.dsavitsk, nikongod, and regal, have given you more than enough good and useful information to actually get started.
.

 

[dBs]

Quote:

As usual, I agree with about 99% of what Doug says. But in the case of his comment about big power tubes,  I'd like to point out that the Japanese DIY community, which historically has been the forefront of the DHT revival, has been running large transmitting tubes at low levels for years. However much this violates generally accepted engineering principals, the results can be surprizingly, even stunningly, good. Sorry to drift off point.
 
dBs, it's definitely time to end the "math gymnastics" and break out the soldering iron.Your making this a lot more complicated than it really is. Most of your problems will be in implementation rather than design.dsavitsk, nikongod, and regal, have given you more than enough good and useful information to actually get started.
.

Yes, but until I have the funds to do that, I may as well learn as much as I can. There is always more to learn!
 

 

[regal]

Quote:

It's a bit of  messy subject. Actual necessary gain and output power for headphone amps is really quite modest. IMO, gain for low Z phones probably should be unity or lower, and for high Z, more like 3 to 4.

I think I posted my gain suggestion but it may have been lost with the transistion.   I had a DAC with a ~2V p-p output (very low but concieveably encountable) with 600 ohm Sextetts an amp with a gain of 7.5x was loud enough with the volume pot all the way maxed when playing 24 bit padded 20 bit HDCD decoded well mastered music(about the worse case scenario.).  Now I have an amp with 10x gain and it is just right for both 32 ohm and the 600 ohm phones.   The newer HDCD software decoders may not have this issue but anyway I think it explains the descrepency between our target design gains.  Also take into account a lot of folks listen to bootleg music (raw soundboards or audience recordings)  where the levels are very low.  The old tapers did this to absolutely avoid clipping when they are recording live music on the fly.   So it can be a messy subject,  I'm probably a worst case scenario with the low source output and the low level bootlegs (believe it are not many of these are better than what the studios put out.)
 
 
 
 

 

[dBs]

Good advice so far, helped give a sense of scope.
 
EDIT: Ha! Well that's ironic. I found the tube I talk about bellow on a site listing different medium-mu tubes. Low and behold I see it was a tube dsavitsk had pointed out. I looked through the ones you pointed, but maybe I accidentally missed/skipped that particular one. Well, you were absolutely right! Good call =D
 
I found a tube that may be more fitting. Of course, my version of "found" at this point may just mean stumbling across one of the most popular tubes, hahaha. In any case its the 5687 dual triode. It seems fairly common, fairly inexpensive, nice and linear from the couple of data sheets I've looked at, has a nice low Rp value between 1650 Ohms @-2Vg and 3000 @-12.5Vg. A mu of 17, same as the 2C22, but because it is a dual triode, I can parallel the two sides together to up the output power so without having to resort to anything exotic, and should allow me to make it a one stage job. Only concern with that tactic is finding an upper tube that can handle the added current that will be needed to run the combined Iq. If need be I can go with a paralleled upper tube as well, but that gets a bit close to http://gilmore2.chem.northwestern.edu/projects/showfile.php?file=barbour_prj.htm and I don't want to bite someone else's design if I can help it.
 
I have to do some number crunching, but at this point, I like what I am seeing in the 5687 tube.

 

[FrankCooter]

If you're looking for a beefier 5687, consider the octal 6bl7.The 6bx7 is even better.
 
The "Brute Force in a Line Stage"  is an exellent first tube DIY project. I've built 4 or 5 different versions. Even one with a DHT (112A) on the bottom.. Worked surprizingly well. This is definitely a good place to start.

 

[Leny]

Quote:

If need be I can go with a paralleled upper tube as well, but that gets a bit close to http://gilmore2.chem.northwestern.edu/projects/showfile.php?file=barbour_prj.htm and I don't want to bite someone else's design if I can help it.
 

Or just copy it and add a Hammond 119DA or (if you are feeling rich) a Sowter 8665 on the output. It would make a good learning project, it would work, and you could progress from there into your first home-grown design!

 

[dBs]

Naaaah, that's one of the reasons I am designing it from the ground up. I want that added challenge, level of customization, and the other flurry of benefits that go along with that method.

 

[dBs]

Ok, I feel confident in my tube choices now.
 
I have the dsavitsk 5687 dual triode lower tube (connected in // for 2 tubes total), and a 12HL7 upper single pentode for the top (for 2 tubes total).
 
The major considerations for the pentode came with the // triode which meant that I needed the pentode to be able to handle the 40mA that I plan to run them at (20mA each since they are in //). I didn't want to have to have a really power PSU so I tried to find a pentode with lower Ec2 and Vp operating points. The pentode also had to be able to accept the voltage swing that the lower 5687 tubes would be using. The 12HL7 on the surface appears to fit the bill nicely.
 
At this point this is how I am looking to operate the 5687:
 
Vg=-3V
Ip=20mA (each)
Vp=107V
Mu=~17
Rp=~1.6-1.7k
 
which yields a Vp-p of 110V
 
I have to crunch some numbers for the pentode which I hope to do tonight, but from what I see, it should work well. I haven't found much on it, but I suspect that the way you set the operating point is as if it was a pentode running in cathode bias cathode follower configuration with the traditional RL in series with Rk. I don't see the lower triodes acting as a CVS so I'm expecting to plan for this using the traditional load line method.

 

[dBs]

Ok, long weekend, hopefully I'll find some time to get this blasted math done. Bit stuck right now finding an RL value. I'll get it, but it has proven less than cooperative, lol.
 
Otherwise, I had a REALLY out there idea to maybe tackle that DHT problem that was mentioned a long time ago in this thread. I think I might have found a way to heat a 300B (for example) without using a transformer of any kind, or a battery of any kind. What about stringing together a series of diodes? Your average solid state diode has a knee voltage of .6-.7V. The 300B has a filament voltage of 5V, that is only 8-9 diodes in series. Now, this will drop your signal by that 5V and that number of diodes in series might produce enough noise to make more of a problem than it solves, but I haven't seen anyone propose that idea before. There may be a reason for that, but I figure I'd propose it =D

 

[nikongod]

What does RL mean?
 
My guess is "load resistance" (Resistance of the Load) as in the headphone, or the reflected load on the primary of the transformer which is not a constant. If I guessed right you are trying to hit a moving target with a stationary gun. In the dark.